Drilling trajectory real-time dynamic tracking control method and system based on deep learning

Abstract

The application discloses a drilling trajectory real-time dynamic tracking control method and system based on deep learning, relates to the technical field of oil drilling, and comprises the following steps: generating a wellbore trajectory control parameter according to basic information of a well and preset wellbore trajectory data; setting optimal drilling fluid performance parameters corresponding to a current working condition according to a stratum environment where a drilling tool is located; performing deep learning on the wellbore trajectory control parameter and the optimal drilling fluid performance parameters through an association model to generate a drilling parameter setting scheme and control the drilling tool to drill a well; comparing drilling data with data in the drilling parameter setting scheme to determine a deviation in the drilling data; determining a corresponding fault mode and correcting the drilling parameter setting scheme by determining a corresponding solution from a solution library. Through deep learning on geological conditions and historical data, the application can continuously update and optimize a model by accumulating data, and continuously improve the drilling trajectory tracking and control precision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil drilling, in particular to a drilling trajectory real-time dynamic tracking control method and system based on deep learning. BACKGROUND

[0002] Drilling trajectory tracking refers to real-time monitoring and control of the drilling path of a drill bit during drilling, so that it is as consistent as possible with the pre-designed well trajectory. In petroleum engineering, drilling trajectory tracking control technology is of great significance to drilling quality, recovery efficiency and capital investment.

[0003] At present, the global oil and gas exploration trend is developing towards ultra-deep water, ultra-deep layer, low permeability, unconventional and other directions, and drilling trajectory tracking control technology is facing great challenges. With the increasing requirements of the oil and gas industry on recovery rate and drilling cost, the traditional drilling trajectory tracking control technology has been unable to meet the needs of complex oil and gas reservoirs in terms of trajectory accuracy and drilling efficiency, and the drilling trajectory tracking control technology needs to be broken through. The research focus of domestic and foreign drilling trajectory tracking control technology mainly concentrates on horizontal wells, directional wells, branch wells and multi-bottom wells, etc. In terms of theory, the research on drilling string mechanics, drilling tool assembly, etc. has been relatively mature, and various types of directional drilling technology has also developed to a certain extent. However, with the increasing complexity of oil and gas reservoirs, drilling trajectory tracking control technology is facing severe challenges brought by high temperature, high pressure and high steepness of complex oil and gas reservoirs. At the same time, the problems of nonlinearity, strong interference, high coupling, hysteresis and time-varying in the drilling process also bring many difficult challenges to the drilling trajectory tracking control technology.

[0004] In the patents CN111810113A and CN110500081A, it is mentioned to establish a rotary steerable drilling database, input real drilling or simulated drilling data; form a drilling model based on the drilling database; input the operation well data in the drilling model, give the operation well drilling tool assembly suggestion, output the operation well drilling scheme; real-time measurement of well trajectory data, evaluation of tool state, identification of well trajectory and design matching degree, planning of well control scheme, division of the latest control instruction drilling section, adjustment of drilling parameters. But both patents do not elaborate on real-time updating of data to the model during drilling.

[0005] With the advancement of drilling, new formation information and drilling conditions are constantly changing. If these new data cannot be timely incorporated into the model training, the accuracy of the model will decrease over time; at the same time, the trained drilling model mixes all drilling data together for training, which may interfere between different types of data, making it difficult for the model to accurately learn the key features of each type of data. SUMMARY

[0006] The drilling trajectory real-time dynamic tracking control method and system based on deep learning provided by the embodiments of the present application are used to solve the problems in the prior art that the drilling model established by the rotary steerable drilling system based on deep learning decreases in accuracy as the formation information and drilling conditions continuously change, the drilling model is difficult to accurately learn the key features of each type of data, and there is no processing measure for abnormal fluctuations in the downhole sensor measurement data.

[0007] In one aspect, the embodiments of the present application provide a drilling trajectory real-time dynamic tracking control method based on deep learning, which comprises the following steps:

[0008] S1, generating borehole trajectory control parameters according to the real-time acquired basic information of the well and the preset borehole trajectory data through a pre-constructed borehole trajectory control model, and setting optimal drilling fluid performance parameters corresponding to the current working condition according to the formation environment in which the drilling tool is located through a pre-constructed drilling fluid formation parameter coupling model;

[0009] S2, performing deep learning on the borehole trajectory control parameters and the optimal drilling fluid performance parameters through a pre-constructed correlation model to generate a drilling parameter setting scheme, and controlling the downhole drilling tool to drill according to the drilling parameter setting scheme;

[0010] S3, comparing the real-time acquired drilling data with the data in the drilling parameter setting scheme through a pre-constructed multi-index fusion evaluation model to determine whether the drilling data deviates;

[0011] S4, determining a fault mode corresponding to the deviation from a pre-constructed fault mode library, and correcting the drilling parameter setting scheme based on the fault mode from a solution library.

[0012] Preferably, before step S1, the method comprises the following steps:

[0013] constructing a deep learning database based on historical data of the well to be drilled, wherein the types of the historical data include one or more of the following: geographical position of the well, well depth, drilling fluid performance parameters, borehole trajectory data, drilling tool type, drilling tool mechanical parameters, basic formation parameters of the well to be drilled, fault type, fault characteristics, and solutions;

[0014] performing data refinement processing on the data in the deep learning database, including data cleaning, data deduplication, data standardization, data anomaly processing, and noise reduction.

[0015] Preferably, the method further comprises the following steps:

[0016] performing deep learning on the borehole trajectory data and the drilling tool mechanical parameters to construct the borehole trajectory control model;

[0017] Deep learning is performed on basic formation parameters of a well to be drilled and performance parameters of a drilling fluid to construct a drilling fluid-formation parameter coupling model.

[0018] Deep learning is performed on well trajectory data and performance parameters of the drilling fluid to construct a correlation model.

[0019] Deep learning is performed on historical data to construct a multi-index fusion evaluation model.

[0020] Preferably, determining that the drilling data deviates in step S3 comprises:

[0021] The real-time acquired well trajectory data exceeds the well trajectory data in the drilling parameter setting scheme; and / or the real-time acquired drilling tool mechanical data exceeds the drilling tool mechanical data in the drilling parameter setting scheme; and / or the real-time acquired drilling fluid performance data exceeds the drilling fluid performance data in the drilling parameter setting scheme.

[0022] Preferably, before step S4, the method further comprises:

[0023] A fault mode library is constructed according to the historical data and experience data, wherein the fault mode library contains: historical fault types, and deviation characteristics corresponding to the historical fault types.

[0024] A solution library is constructed according to a pre-set rule library and an optimization algorithm.

[0025] Determining the fault mode corresponding to the deviation in step S4 comprises:

[0026] Matching the characteristics of the deviation with the fault modes in the fault mode library to identify the most similar fault mode.

[0027] Based on the fault mode, the corresponding solution is obtained from the solution library.

[0028] Preferably, the method further comprises: taking the real-time acquired drilling data as a new sample to update parameters of the well trajectory control model and the drilling fluid-formation parameter coupling model.

[0029] In another aspect, the embodiments of the present application also provide a drilling trajectory real-time dynamic tracking control system based on deep learning, which comprises:

[0030] A database for storing historical data of a well region to be drilled;

[0031] A model training module for deep learning on data in the database to construct a well trajectory control model, a drilling fluid-formation parameter coupling model, a multi-index fusion evaluation model, and a correlation model by deep learning on data parameters output by the well trajectory control model and the drilling fluid-formation parameter coupling model;

[0032] The drilling parameter generation module is configured to generate a drilling parameter setting scheme for the well based on basic information of the well, preset well trajectory data, and a formation environment in which a drilling tool is located, by using a well trajectory control model, a drilling fluid formation parameter coupling model, and a correlation model.

[0033] The drilling data verification module is configured to compare the real-time acquired drilling data with data in the drilling parameter setting scheme to determine whether a deviation occurs.

[0034] The drilling parameter optimization module is configured to optimize the drilling parameter setting scheme by using a corresponding solution when the drilling data deviates.

[0035] Preferably, the drilling parameter verification module comprises:

[0036] The real-time monitoring unit is configured to acquire the drilling data in real time.

[0037] The comparison unit is configured to determine that the drilling data deviates when the drilling data exceeds parameters in the drilling parameter setting scheme.

[0038] Preferably, the drilling parameter optimization module comprises:

[0039] The fault mode matching unit is configured to match the deviation feature with fault modes in a preset fault mode library, identify the most similar fault mode, and obtain a corresponding solution from a preset solution library based on the fault mode.

[0040] The parameter optimization unit is configured to fuse the solution into the drilling parameter setting scheme for correction.

[0041] In another aspect, the present application also provides a computer readable storage medium, which stores a computer program. When the computer program is executed, the computer program implements the above-mentioned deep learning-based drilling trajectory real-time dynamic tracking control method.

[0042] The deep learning-based drilling trajectory real-time dynamic tracking method and system has the following advantages:

[0043] By deep learning training on complex and variable geological conditions, a large amount of historical multi-source data from different sensors and geology, and the like, well trajectory control models, drilling fluid formation parameter coupling models, multi-index fusion evaluation models, and correlation models are constructed, and logging data is fed back in real time to enable the processing models to have self-learning ability, continuously update and optimize the models as data accumulates, and continuously improve the drilling trajectory tracking and control precision without the need for frequent reprogramming or adjustment of algorithms by humans. Meanwhile, the multi-index fusion evaluation model provides reliable guidance for drilling operations, can simultaneously consider and comprehensively analyze and decide multiple data states such as well trajectory, drilling tool equipment state, and drilling fluid performance, and develop a more scientific and reasonable trajectory tracking scheme to improve overall efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0045] Figure 1 A flowchart of a method for real-time dynamic tracking of drilling trajectories based on deep learning provided in an embodiment of the present application.

[0046] Figure 2 A schematic diagram of the structure of a real-time dynamic tracking system for drilling trajectories based on deep learning provided in an embodiment of the present application.

[0047] Figure 3 Schematic diagram of a real-time dynamic tracking system for drilling trajectories based on deep learning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] refer to Figure 1 The figure shows a flow chart of a method for real-time dynamic tracking of drilling trajectories based on deep learning provided by an embodiment of the present application. The method for real-time dynamic tracking and control of drilling trajectories based on deep learning provided by an embodiment of the present application includes:

[0050] S1. Generate wellbore trajectory control parameters based on the real-time acquired well basic information and preset wellbore trajectory data through a pre-built wellbore trajectory control model, and set the optimal drilling fluid performance parameters corresponding to the current working conditions based on the formation environment where the drill tool is located through a pre-built drilling fluid formation parameter coupling model.

[0051] In this embodiment, before starting drilling, the well trajectory control parameters are first produced by the pre-constructed well trajectory control model according to the real-time acquired basic information of the well, such as the current well depth, formation lithology, wellbore temperature, etc., and the preset well trajectory data, such as the target inclination angle, azimuth angle, target point coordinates, etc. The well trajectory control parameters include, for example, the drilling pressure adjustment value, the deflection angle of the steering tool, the rotation speed optimization suggestion, etc. When the drilling tool enters a high-pressure formation, the drilling fluid density needs to be increased to balance the formation pressure and prevent blowout or well wall collapse. Therefore, the optimal drilling fluid performance parameters corresponding to the current working condition are set by the pre-constructed drilling fluid formation parameter coupling model according to the current bottom-hole environment of the drilling tool, such as the current formation lithology, porosity, fracture pressure, etc., and according to the density and viscosity range set according to the formation stability requirement. For example, the drilling fluid density is increased to 1.25 g / cm³, and the viscosity is adjusted to 45 mPa·s.

[0052] S2, depth learning of the well trajectory control parameters and the optimal drilling fluid performance parameters is performed by the pre-constructed correlation model to generate a drilling parameter setting scheme. The drilling parameter setting scheme includes, but is not limited to, the well trajectory control instruction, the drill bit mechanical parameter setting instruction, the working parameter of the drilling pump, and the drilling fluid performance parameter.

[0053] In this embodiment, the well trajectory control parameters, such as the drilling pressure adjustment value, the deflection angle of the steering tool, and the rotation speed optimization suggestion, and the optimal drilling fluid performance parameters, such as the drilling fluid density and the viscosity, are subjected to depth learning data analysis by the pre-constructed correlation model to form a drilling parameter setting scheme. The drilling parameter setting scheme includes, for example, the well trajectory control instruction, such as the drilling pressure maintaining 85 kN and the tool face angle adjusting to 12°; the drill bit mechanical parameter setting instruction, such as the rotation speed increasing to 120 RPM (revolutions per minute) and the torque limit being 50 kN·m; the working parameter of the drilling pump, such as the pump pressure maintaining 20 MPa and the displacement adjusting to 1.2 m³ / min; and the drilling fluid performance parameter, such as the density 1.2 g / cm³, the viscosity 40 mPa·s, and the shear force 4 Pa.

[0054] S3, the real-time acquired drilling data and the data in the drilling parameter setting scheme are compared by the pre-constructed multi-index fusion evaluation model to determine whether the drilling data deviates.

[0055] In the present embodiment, during drilling, real-time measurement of current well trajectory data, drilling tool mechanical data, and drilling fluid performance data, etc. drilling data is also needed to comprehensively judge whether the current drilling data deviates, for example: through the measurement while drilling (MWD) tool to obtain accurate well inclination angle, azimuth angle, tool face angle, etc. data, and real-time well depth information recorded by the well depth measuring sensor; through the geosteering instrument and logging equipment to collect the parameters of the formation lithology, porosity, permeability, formation pressure, etc.; through the drilling fluid performance monitoring instrument to obtain the drilling fluid density measured by the densimeter, the drilling fluid viscosity measured by the viscometer, the drilling fluid shear force detected by the shear force meter, and the drilling fluid loss measured by the filter loss instrument, etc.; through the drilling tool mechanical data sensor to obtain the drilling pressure data measured by the pressure sensor installed on the drilling tool, the drilling tool rotation speed recorded by the rotation speed sensor, and the torque size borne by the drilling tool monitored by the torque sensor. In the present embodiment, before comparison, the collected data also needs to be cleaned to remove noise, outliers and missing values; different types and ranges of data are normalized to have the same scale.

[0056] By comparing the real-time measured current well trajectory data, drilling tool mechanical data, and drilling fluid performance data, etc. with the data in the drilling parameter setting scheme output by the above correlation model, when the real-time measured data is inconsistent with the data in the drilling parameter setting scheme or the difference reaches a certain threshold, it is judged that the drilling data deviates.

[0057] In the present embodiment, the well inclination angle, azimuth angle, well depth, etc. well trajectory data obtained by real-time measurement are compared with the parameters in the well trajectory control instruction output by the correlation model, for example: when the difference between the actual well inclination angle and the designed well inclination angle exceeds the specified error range (for example ±0.5°), or the actual azimuth angle deviates from the designed azimuth angle to a certain extent (such as ±1°), it can be judged that the well trajectory data deviates.

[0058] In the embodiment, the real-time measured drilling mechanical data such as WOB, RPM, torque, etc. are compared with the parameters in the drilling mechanical parameter setting instruction output by the correlation model. For example, when the WOB suddenly fluctuates greatly and deviates from the normal working range (for example, the set WOB is 80-100 kN, and the actual WOB is less than 60 kN or higher than 120 kN momentarily); or the torque abnormally increases and exceeds the normal bearing range of the drilling tool (for example, the normal torque is 30-50 kN·m, and the actual torque reaches more than 70 kN·m), it indicates that the drilling tool working state is abnormal, and the drilling data deviates. In the embodiment, the real-time measured drilling fluid performance data such as density, viscosity, shear force, etc. are compared with the standard value parameters in the drilling fluid performance parameters determined according to the formation conditions and drilling process requirements. For example, when the deviation of the drilling fluid density from the standard density exceeds ±0.05 g / cm³, the viscosity fluctuation exceeds ±5 mPa·s, or the shear force changes obviously (for example, the standard shear force is 3-5 Pa, and the actual shear force is less than 2 Pa or higher than 6 Pa), it indicates that the drilling fluid performance data deviates.

[0059] S4, if yes, determine the fault mode corresponding to the deviation, and modify the drilling parameter setting scheme with the solution corresponding to the fault mode.

[0060] In the embodiment, after determining that the drilling data deviates, the most similar fault mode is found out by comparing the type, amplitude, change trend, etc. of the deviation, and then a suitable solution is selected from the pre-prepared solution library, and the solution is integrated into the drilling parameter setting scheme to adjust the drilling parameters and correct the problems in real time.

[0061] In the embodiment of the present application, before starting the drilling operation, a database for deep learning is constructed based on the historical well logging data, simulated drilling data, historical geological exploration data and historical fault information of the to-be-drilled well area, wherein the constructed database at least contains: the geographical position of the well, the well depth, the drilling fluid performance parameters, the well trajectory data, the drilling tool type, the drilling tool mechanical data and the basic formation parameters of the to-be-drilled well, etc. The drilling fluid performance parameters include: the density, viscosity, PH value, temperature and conductivity of the drilling fluid, which can be obtained by the drilling fluid performance monitoring system established during the drilling process. The drilling tool mechanical parameters include: the WOB applied on the drill bit, the rotation speed of the drill string and the drill bit, and the torque when the drill string rotates. The well trajectory data includes: the inclination angle, the azimuth angle and the tool face angle. The drilling tool type includes: the push-type rotary steerable drilling tool, the pointing-type rotary steerable drilling tool, etc. The basic formation parameters of the to-be-drilled well include: the gamma ray data, the resistivity data, the formation pore pressure data, the formation rock fracture pressure data; the fault information includes: the fault type, the fault characteristics and the solution, etc.

[0062] In this embodiment, in order to enable the data in the deep learning database to facilitate the training of the deep learning model, data refinement processing such as data cleaning, data deduplication, data standardization, data anomaly processing, and noise reduction needs to be performed on the data in the database. For example, the data of the geographic position of the well, the drilling tool mechanical parameters, and the wellbore trajectory in the database are checked for integrity, and missing values are filled. For a small amount of missing values, a suitable interpolation method can be selected according to the characteristics of the data. The data of different characteristics are standardized to have similar scales and distributions.

[0063] In the embodiments of the present application, the wellbore trajectory parameters, the drilling tool type, and the drilling tool mechanical parameters are trained by deep learning to obtain a wellbore trajectory control model; the drilling fluid performance parameters and the basic formation parameters of the well to be drilled are trained by deep learning to obtain a drilling fluid formation parameter coupling model; and all data are analyzed by learning to obtain a multi-index fusion evaluation model. In this embodiment, for the wellbore trajectory control model and the drilling fluid formation parameter coupling model, an incremental learning method is used for updating. When new real-time data arrives, the model adjusts the parameters of the model to adapt to the new data characteristics on the basis of the original knowledge. For example, the drilling data obtained in real time are used as new samples to update the parameters of the wellbore trajectory control model and the drilling fluid formation parameter coupling model.

[0064] In the embodiments of the present application, the optimal control strategy and control instruction calculated based on the expected wellbore trajectory, including the drilling pressure, the rotation speed, and the steering tool parameters, obtained from the wellbore trajectory control model, and the drilling fluid performance parameters, such as the density, the viscosity, and the shear force, dynamically adjusted according to the real-time formation parameters, obtained from the drilling fluid formation parameter coupling model, are analyzed and deep learning is performed to establish a correlation model between the wellbore trajectory control parameters and the drilling fluid performance parameters, and to quantify the degree of mutual influence therebetween. According to the target of requiring real-time dynamic tracking and control of the drilling trajectory in the drilling operation, a comprehensive objective function is determined; the objective function weights multiple indexes such as the rate of penetration, the drilling fluid parameters, and the wellbore trajectory deviation; and an optimization algorithm is used to comprehensively optimize the wellbore trajectory control parameters and the drilling fluid performance parameters under the premise of meeting various constraint conditions such as the load-carrying capacity of the drilling tool and the formation pressure limit, so as to solve a set of optimal drilling parameter combinations.

[0065] According to the optimized drilling parameter combinations, a detailed drilling parameter setting scheme is formulated, including the specific parameter values of the drilling pressure, the rotation speed, the steering tool setting, the drilling fluid density, the viscosity, the shear force, and the adjustment range and adjustment timing of the parameters.

[0066] In the embodiment of the present application, a fault mode library is also constructed according to historical data and empirical data, wherein the fault mode library contains: historical fault types, and deviation characteristics corresponding to the historical fault types; and a solution library is constructed according to a pre-set rule library and an optimization algorithm. In the embodiment, in order to facilitate finding a suitable solution through a fault mode, data in the fault mode library and data in the solution library are also associated, so that when it is determined that the real-time acquired drilling data deviates, a corresponding solution can be found and acquired according to the characteristics of the deviation.

[0067] With reference to Figure 2 and Figure 3 The drilling trajectory real-time dynamic tracking control system based on deep learning provided in the embodiment of the present application comprises: a database, a model training module, a drilling parameter generation module, a drilling data verification module, and a drilling parameter optimization module.

[0068] In the embodiment of the present application, the system stores historical data of a to-be-drilled well region, such as historical well logging data, simulated drilling data, historical geological exploration data, and historical fault information, into the database for subsequent deep learning. In the embodiment, the stored data should contain the geographical position of the well, the well depth, drilling fluid performance parameters, borehole trajectory data, drilling tool types, drilling tool mechanical parameters, and basic formation parameters of the to-be-drilled well, wherein the drilling fluid performance parameters include: the density, viscosity, PH value, temperature, and conductivity of the drilling fluid, which can be acquired through a drilling fluid performance monitoring system established during drilling. The drilling tool mechanical parameters include: the drilling pressure applied on the drill bit, the rotation speed of the drill string and the drill bit, and the torque when the drill string rotates. The borehole trajectory data includes: the inclination angle, the azimuth angle, and the tool face angle. The drilling tool types include: push-against rotary steerable drilling tools, pointing rotary steerable drilling tools, etc. The basic formation parameters of the to-be-drilled well include: gamma ray data, resistivity data, formation pore pressure data, and formation rock fracture pressure data; and the fault information includes: fault types, fault characteristics, and solutions, etc. In the embodiment, after the historical data is stored into the database, the geographical position of the well, the drilling tool mechanical parameters, and the borehole trajectory data in the database are also checked for completeness, and missing values are filled. For a small amount of missing values, a suitable interpolation method can be selected according to the characteristics of the data. The data of different characteristics are standardized to have similar scales and distributions.

[0069] The system performs deep learning on the data stored in the database through the model training module, and classifies and trains models for the mechanical data of the drilling tool, the borehole trajectory control data, the drilling fluid performance parameters, and the formation parameters according to the different types of parameters that need to be set in the rotary steerable drilling system.

[0070] In an embodiment, the wellbore trajectory parameters, drill string type, and drill string mechanical parameters are taken as input features, a wellbore trajectory control model is constructed through deep learning training, and the model can quickly calculate wellbore trajectory control parameters according to the expected wellbore trajectory.

[0071] In an embodiment, the drilling fluid performance parameters and basic formation parameters of the well to be drilled are trained through deep learning to obtain a drilling fluid formation parameter coupling model, and the model can dynamically adjust the drilling fluid performance parameters according to the real-time acquired formation parameters, so that the drilling fluid performance is real-time adapted to the formation conditions.

[0072] In an embodiment, a multi-index fusion evaluation model is obtained by learning all historical data, real-time acquisition of various data such as wellbore trajectory parameters, drill string mechanical parameters, and drilling fluid performance parameters is performed, the data is input into the evaluation model, the model can timely evaluate the drilling condition, and provide decision-making for optimization of drilling parameters. In the embodiment, the multi-index fusion evaluation model is also used to establish a fault mode library according to historical data and experience.

[0073] In an embodiment, the optimal control strategy and control instruction calculated based on the expected wellbore trajectory, including the drilling pressure, rotation speed, and steering tool parameters, are acquired from the wellbore trajectory control model, the drilling fluid performance parameters dynamically adjusted according to the real-time formation parameters, such as the density, viscosity, and shear force, are acquired from the drilling fluid formation parameter coupling model, data analysis and deep learning are performed, a correlation model between the wellbore trajectory control parameters and the drilling fluid performance parameters is established, the model is used to quantify the mutual influence degree between the wellbore trajectory control parameters and the drilling fluid performance parameters, the model determines a comprehensive objective function according to the target of real-time dynamic tracking and control of the drilling trajectory in the drilling operation; the objective function performs weighting and summing on multiple indexes such as the rate of penetration, drilling fluid parameters, and wellbore trajectory deviation; an optimization algorithm is used to comprehensively optimize the wellbore trajectory control parameters and the drilling fluid performance parameters under the premise of meeting various constraint conditions such as the drill string carrying capacity and formation pressure limitation, and a set of optimal drilling parameter combinations is solved. According to the optimized drilling parameter combinations, a detailed drilling parameter setting scheme is formulated, including specific parameter values such as the drilling pressure, rotation speed, steering tool setting, drilling fluid density, viscosity, shear force, and adjustment range and adjustment timing of the parameters.

[0074] The system generates a drilling parameter setting scheme for drilling through the drilling parameter generation module. First, the wellbore trajectory control model obtained above is run to quickly calculate the wellbore trajectory control parameters according to the expected wellbore trajectory; the drilling fluid formation parameter coupling model obtained above is run to dynamically adjust the drilling fluid performance parameters according to the real-time acquired formation parameters; and the drilling parameter setting scheme is obtained from the correlation model according to the wellbore trajectory control parameters and the drilling fluid performance parameters.

[0075] The system runs a multi-index fusion evaluation model through a drilling data verification module, compares the real-time acquired drilling data with the data in the above drilling parameter setting scheme, and determines whether a deviation occurs. In this embodiment, the deviation value is calculated between the real-time well trajectory related data such as the inclination angle, the azimuth angle, and the tool face angle and the set well trajectory. The smaller the deviation value is, the closer the current well trajectory is to the designed trajectory. The three-dimensional coordinates of the drill bit at the bottom of the well (usually using a measurement-while-drilling instrument) are compared with the coordinates of the target point, the distance between the two is calculated, and the closer the distance is, the higher the target accuracy is. In this embodiment, the working efficiency of the drilling tool is evaluated by analyzing the drilling tool mechanical parameters, for example, the torque of the drilling tool is too large, the power consumption is abnormal, and the drilling process is not smooth, which can be used to diagnose the drilling problem and further optimize the well trajectory control. In this embodiment, the fluctuation of the drilling fluid performance parameters such as the density, the viscosity, and the shear force of the drilling fluid in the drilling process is evaluated, and the matching degree of the drilling fluid performance and the real-time formation parameters is evaluated. The drilling fluid with good adaptability can effectively reduce the adverse effects of the formation on the well trajectory, such as in the easy-caving formation, the appropriate drilling fluid density and viscosity can prevent the well wall from collapsing and ensure the smooth extension of the well trajectory.

[0076] The system, through a drilling parameter optimization module, matches the current deviation characteristics with various fault modes in the fault mode library when the drilling data deviates. By comparing the type, amplitude, and change trend of the deviation, the most similar fault mode is found. Further, a suitable solution is selected from the pre-prepared solution library according to the fault mode. The solution is fused into the drilling parameter setting scheme, the drilling parameters are adjusted, and the problem is corrected in real time.

[0077] In this embodiment, after the most similar fault mode is found, a preliminary decision suggestion can also be generated based on the pre-set rules and expert knowledge, and sent to the control system of the drilling equipment and the ground operating personnel, to automatically adjust the drilling parameter setting scheme. After the parameter adjustment, the changes of the related indexes are continuously evaluated according to the real-time feedback data, and the implementation effect of the decision is evaluated. If the deviation still exists or the expected improvement effect is not achieved, the problem diagnosis and analysis are continued, and the decision is re-adjusted.

[0078] In the embodiments of the present application, the drilling data verification module comprises a real-time monitoring unit and a comparison unit.

[0079] In the embodiment, the drilling data verification module acquires drilling data in real time through the real-time monitoring unit, for example, acquires inclination, azimuth, depth and other data of the well trajectory through a measurement while drilling (MWD) tool; acquires parameters such as lithology, porosity, permeability, and formation pressure of the formation through a geosteering instrument and logging equipment; and measures data such as density, viscosity, shear force, and filtration loss of the drilling fluid in real time with the aid of a drilling fluid performance monitoring instrument. The acquired drilling data are compared with the drilling parameter setting scheme through the comparison unit. When the well trajectory data in the drilling data deviates from the preset well trajectory data, for example, when the drilling data deviates from the drilling parameter setting scheme, and when the drilling fluid performance data exceeds the standard drilling fluid pressure performance data, it is determined that the drilling data deviates.

[0080] In the embodiment, the drilling data verification module further includes an updating unit for cleaning the drilling data acquired by the real-time monitoring unit, such as removing noise, outliers, and missing values, and normalizing different types and ranges of data to have the same scale. The incremental learning method is adopted to adapt the new data features by adjusting the parameters of the well trajectory control model and the drilling fluid formation parameter coupling model on the basis of the original knowledge.

[0081] In the embodiment, the drilling parameter optimization module includes a fault mode matching unit and a parameter optimization unit.

[0082] In the embodiment, the drilling parameter optimization module matches the deviation features with fault modes in a preset fault mode library through the fault mode matching unit, identifies the most similar fault mode, and acquires the corresponding solution from a preset solution library based on the fault mode. The solution is fused into the drilling parameter setting scheme by the parameter optimization unit for correction. In the embodiment, after the most similar fault mode is determined through the fault matching unit, a preliminary decision suggestion is generated according to the pre-set rules and expert knowledge through the expert rule library, and is sent to the control system of the drilling equipment and the ground operator to automatically adjust the drilling parameter setting scheme.

[0083] The embodiment of the present application also provides a computer readable storage medium, and the storage medium stores a computer program. The computer program is executed to implement the drilling trajectory real-time dynamic tracking control method based on deep learning.

[0084] In summary, the application uses a deep learning model to train complex and variable geological conditions, a large amount of data from different sensors and geology, and real-time feedback of logging data to make the model have autonomous learning ability, which can continuously update and optimize the model with data accumulation, continuously improve the drilling trajectory tracking and control accuracy, and does not need to frequently reprogram or adjust the algorithm manually. At the same time, training a multi-index fusion evaluation model provides reliable guidance for drilling operations, which can simultaneously consider well trajectory, drilling equipment state, drilling fluid performance and other data states and make comprehensive analysis and decision, and make a more scientific and reasonable trajectory tracking plan to improve overall efficiency.

[0085] Although preferred embodiments of the application have been described, those skilled in the art will be able to make additional changes and modifications to these embodiments once they have been given the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the application.

[0086] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A real-time dynamic tracking control method for drilling trajectory based on deep learning, characterized in that: The method comprises: S1. Generate wellbore trajectory control parameters based on real-time acquired well basic information and preset wellbore trajectory data using a pre-built wellbore trajectory control model; and set optimal drilling fluid performance parameters corresponding to the current operating conditions based on the formation environment in which the drilling tool is located using a pre-built drilling fluid formation parameter coupling model; S2. Deeply learning the wellbore trajectory control parameters and the optimal drilling fluid performance parameters using a pre-built correlation model to generate a drilling parameter setting plan, and controlling the downhole drilling tool to drill according to the drilling parameter setting plan; S3. Comparing the real-time acquired drilling data with the data in the drilling parameter setting plan using a pre-built multi-index fusion evaluation model to determine whether the drilling data has any deviation; S4. Determine a fault mode corresponding to the deviation from a pre-built fault mode library, and determine a corresponding solution from a solution library based on the fault mode to modify the drilling parameter setting solution; Wherein, before step S1, the following steps are included: Constructing a deep learning database based on historical data of the area to be drilled, wherein the types of historical data include one or more of the following: geographic location of the well, well depth, drilling fluid performance parameters, wellbore trajectory data, drilling tool type, drilling tool mechanical parameters, basic formation parameters of the well to be drilled, fault type, fault characteristics, and solutions; The method further comprises: performing deep learning on the wellbore trajectory data and the drilling tool mechanical parameters to construct the wellbore trajectory control model; Performing deep learning on the basic formation parameters of the well to be drilled and the drilling fluid performance parameters to construct a drilling fluid formation parameter coupling model; performing deep learning on the wellbore trajectory data and the drilling fluid performance parameters to construct the correlation model; Performing deep learning on the historical data to construct the multi-index fusion evaluation model; Determining that the drilling data has a deviation in step S3 includes: The wellbore trajectory data acquired in real time exceeds the wellbore trajectory data in the drilling parameter setting plan; and / or the drill tool mechanical data acquired in real time exceeds the drill tool mechanical data in the drilling parameter setting plan; and / or the drilling fluid performance data acquired in real time exceeds the drilling fluid performance data in the drilling parameter setting plan.

2. The real-time dynamic tracking control method for drilling trajectory based on deep learning according to claim 1, characterized in that: The data in the deep learning database is subjected to data refinement processing, including data cleaning, data deduplication, data standardization, data anomaly processing and noise reduction.

3. The real-time dynamic tracking control method for drilling trajectory based on deep learning according to claim 1, characterized in that: Before step S4, the method further includes: Constructing a fault mode library based on historical data and empirical data, wherein the fault mode library includes: historical fault types and deviation features corresponding to the historical fault types; Build a solution library based on pre-set rule base and optimization algorithm; Determining the failure mode corresponding to the deviation in step S4 includes: Matching the characteristics of the deviation with the failure modes in the failure mode library to identify the most similar failure mode; A corresponding solution is obtained from the solution library based on the failure mode.

4. The real-time dynamic tracking control method for drilling trajectory based on deep learning according to claim 1, characterized in that: The method further comprises: The real-time acquired drilling data is used as a new sample to update the parameters of the wellbore trajectory control model and the drilling fluid formation parameter coupling model.

5. A real-time dynamic tracking and control system for drilling trajectories based on deep learning, characterized in that: The system applies the method according to claim 1, and the system includes: A database for storing historical data of the area to be drilled; a model training module for performing deep learning on the data in the database to construct a wellbore trajectory control model, a drilling fluid formation parameter coupling model, and a multi-index fusion evaluation model, and for performing deep learning on the data parameters output by the wellbore trajectory control model and the drilling fluid formation parameter coupling model to construct an association model; A drilling parameter generation module is used to generate a drilling parameter setting plan based on the basic information of the well, the preset well trajectory data and the formation environment where the drilling tool is located through the well trajectory control model, the drilling fluid formation parameter coupling model and the association model; A drilling data verification module is used to compare the drilling data acquired in real time with the data in the drilling parameter setting plan to determine whether there is a deviation; The drilling parameter optimization module is used to obtain a corresponding solution to optimize the drilling parameter setting solution when deviation occurs in the drilling data.

6. The real-time dynamic tracking and control system for drilling trajectory based on deep learning according to claim 5, characterized in that: The drilling data verification module includes: Real-time monitoring unit, used to obtain drilling data in real time; The comparison unit is configured to determine that a deviation occurs in the drilling data when the drilling data exceeds a parameter in the drilling parameter setting plan.

7. The real-time dynamic tracking and control system for drilling trajectory based on deep learning according to claim 5, characterized in that: The drilling parameter optimization module includes: A fault pattern matching unit is used to match the deviation characteristics with the fault patterns in a preset fault pattern library, identify the most similar fault pattern, and obtain a corresponding solution from a preset solution library based on the fault pattern; A parameter optimization unit is used to integrate the solution into the drilling parameter setting plan for correction.

8. A computer-readable storage medium, characterized in that The storage medium stores a computer program, which, when executed, implements the real-time dynamic tracking and control method for drilling trajectories based on deep learning as described in any one of claims 1 to 4.

Images